Review © American College of Medical and Open

Implementing genomic in the clinic: the future is here

Teri A. Manolio, MD, PhD1, Rex L. Chisholm, PhD2, Brad Ozenberger, PhD1, Dan M. Roden, MD3, Marc S. Williams, MD4,5, Richard Wilson, PhD6, David Bick, MD7, Erwin P. Bottinger, MD8, Murray H. Brilliant, PhD9, Charis Eng, MD, PhD10, Kelly A. Frazer, PhD11, Bruce Korf, MD, PhD12, David H. Ledbetter, PhD5, James R. Lupski, MD, PhD13, Clay Marsh, MD14, David Mrazek, MD15, Michael F. Murray, MD16, Peter H. O’Donnell, MD17, Daniel J. Rader, MD18, Mary V. Relling, PharmD19, Alan R. Shuldiner, MD20, David Valle, MD21, Richard Weinshilboum, MD22, Eric D. Green, MD, PhD1 and Geoffrey S. Ginsburg, MD, PhD23

Although the potential for genomics to contribute to clinical care relevant; lack of reimbursement for genomically driven interventions; has long been anticipated, the pace of defining the risks and benefits and burden to patients and clinicians of assaying, reporting, inter- of incorporating genomic findings into medical practice has been vening, and following up genomic findings. Key infrastructure needs relatively slow. Several institutions have recently begun genomic included an openly accessible knowledge base capturing sequence medicine programs, encountering many of the same obstacles and variants and their phenotypic associations and a framework for developing the same solutions, often independently. Recognizing defining and cataloging clinically actionable variants. Multiple insti- that successful early experiences can inform subsequent efforts, the tutions are actively engaged in using genomic information in clinical National Research Institute brought together a care. Much of this work is being done in isolation and would benefit number of these groups to describe their ongoing projects and chal- from more structured collaboration and sharing of best practices. lenges, identify common infrastructure and research needs, and out- line an implementation framework for investigating and introducing Genet Med 2013:15(4):258–267 similar programs elsewhere. Chief among the challenges were limited evidence and consensus on which genomic variants were medically Key Words: medical genomics; practice standards

INTRODUCTION individual patient’s genotypic information in his or her clinical The potential for genomics to contribute to clinical care has care. This definition encompasses both Mendelian and multi- long been recognized, and many optimistic scenarios for clini- genic complex diseases, and at present probably focuses most cal use of information about a patient’s genome have been on single Mendelian variants of large effect, although emphasis proposed.1–3 The pace of realizing this potential has appeared is expected to shift soon to assaying and using multiple vari- slow to some,4,5 although clinical adoption of scientific discov- ants simultaneously in clinical care. This is not only because eries has been estimated to take up to 17 years6 and a recent of an expanding knowledge base but also because the logistic genetic example7 required 18 years after the initial 1991 report.8 complexities and costs of assessing genomic variation on a large Indeed, relatively robust genotype–phenotype associations for scale are approaching those of one-at-a-time testing of individ- common, complex diseases only began to become available ual variants, permitting a more holistic approach to incorporat- around 2005.9 Yet several academic medical centers and inte- ing genomic findings into clinical care. grated health systems have already begun programs for imple- These initial and separate forays have encountered similar menting genomic medicine, which we define here as using an obstacles and developed many of the same nascent solutions,

1National Research Institute, National Institutes of Health, Bethesda, Maryland, USA; 2Center for Genetic Medicine, Northwestern University, Chicago, Illinois, USA; 3Departments of Medicine and Pharmacology, Vanderbilt University School of Medicine, Nashville, Tennessee, USA; 4Clinical Genetics Institute, Intermountain Healthcare, Salt Lake City, Utah, USA; 5Genomic Medicine Institute, Geisinger Health System, Danville, Pennsylvania, USA; 6The Genome Institute, Washington University School of Medicine, St. Louis, Missouri, USA; 7Division of Genetics, Department of , Medical College of Wisconsin, Milwaukee, Wisconsin, USA; 8 The Charles Bronfman Institute for , Mount Sinai School of Medicine, New York, New York, USA; 9Center for Human Genetics, Marshfield Clinic, Marshfield, Wisconsin, USA; 10Genomic Medicine Institute, Cleveland Clinic Lerner Research Institute, Cleveland, Ohio, USA; 11Department of Pediatrics, Division of Genome Information Sciences, University of California San Diego, La Jolla, California, USA; 12Department of Genetics, University of Alabama at Birmingham, Birmingham, Alabama, USA; 13Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, Texas, USA; 14Center for Personalized Health Care, Ohio State University Medical Center, Columbus, Ohio, USA; 15Department of and , Mayo Clinic, Rochester, Minnesota, USA; 16Genetics Division, Brigham and Women’s Hospital, Boston, Massachusetts, USA; 17Department of Medicine, University of Chicago Medical Center, Chicago, Illinois, USA; 18Division of Translational Medicine and Human Genetics, Department of Medicine, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, Pennsylvania, USA; 19 Department of Pharmaceutical Sciences, St. Jude Children’s Research Hospital, Memphis, Tennessee, USA; 20Program in Personalized and Genomic Medicine, Department of Medicine, University of Maryland School of Medicine, Baltimore, Maryland, USA; 21Institute of Genetic Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland, USA; 22Department of Molecular Pharmacology and Experimental Therapeutics, Mayo Clinic College of Medicine, Rochester, Minnesota, USA; 23Genomic Medicine, Duke Institute for Genome Sciences & Policy; Center for Personalized Medicine, Duke University, Durham, North Carolina, USA. Correspondence: Teri A. Manolio ([email protected])

Submitted 16 April 2012; accepted 30 October 2012; advance online publication 10 January 2013. doi: 10.1038/gim.2012.157

258 Volume 15 | Number 4 | April 2013 | Genetics in medicine Implementing genomic medicine in the clinic | MANOLIO et al Review often quite independently. In expectation that sharing of les- approaches. More important, perhaps, successful implemen- sons learned in these efforts can facilitate broader and more tation of screening programs at early adopter sites provides effective implementation of genomic medicine, the National valuable lessons and refined approaches that may be adopted Human Genome Research Institute in collaboration with sev- elsewhere.13 eral leaders in this area (R.L.C., G.S.G., D.M.R., M.S.W., and Self-reported family history information can also be used R.Wilson) brought together about 20 groups working to imple- in risk assessment for individual patients with documented ment genomic medicine for a Genomic Medicine Colloquium clinical validity and utility,14,15 despite its potential weaknesses in June 2011. This article summarizes the projects described such as incomplete or inaccurate reporting.16 Family history by these groups, describes challenges they have encountered is generally not difficult to collect, especially with electronic in implementation, identifies common infrastructure and data collection tools such as the Surgeon General’s My Family research needs, and outlines an implementation framework for Health Portrait,17 although integrating the information into an introducing similar genomic medicine programs more broadly. independent electronic medical record (EMR) is not trivial. Clinicians are generally familiar with family history and appre- BRIEF LANDSCAPE OF GENOMIC MEDICINE ciate its importance, and evidence-based guidelines are avail- PROJECTS able for several conditions.12,14 A Duke University pilot proj- Participating sites reported a broad range of genomic medicine ect, for example, is collecting three-generation family history activities, both in pilot and full implementation forms (Table 1). information on 48 diseases using a Web-based computerized These included (used here and throughout to refer tool.18 Patient-entered information is integrated into the medi- to the detection of genetic variants by single- poly- cal record and generates a pedigree, tabular family history, and morphism assay, , or other technologies) of somatic reports for the patient and clinician, as well as decision support in malignant tumors to guide treatment decisions. information for four conditions (breast, ovarian, and colorec- The expanding scope and growing acceptance of such tech- tal cancer and venous thrombosis). Outcomes being assessed niques places tumor genotyping at the forefront of genotype- include patient and clinician satisfaction and ease of use; directed care. All the participating sites (and most cancer cen- changes in patient behaviors such as diet and smoking; changes ters nationwide) conduct tumor genotyping of cancers such as in behaviors such as screening and referral; and esti- melanoma and those of the breast, colon, and lung for the tar- mates of sensitivity, specificity, net reclassification, and cost.18 geting of . Several also use genotyping to select patients Similarly, a Cleveland Clinic prototype collects three-genera- for interventional protocols, but the main focus here was on tion family histories for the assessment of all inherited cancer the use of genomics outside of cancer diagnosis and treatment. syndromes.19 Intermountain Healthcare has also deployed a Several centers have begun targeted screening for highly patient-entered family tool (“Our Family Health”) in their elec- penetrant germline mutations for Lynch syndrome and in tronic patient portal.20 BRCA1/210–12 to identify genetically at-risk individuals. With provides additional opportunities in appropriate immunohistochemical or other evidence sugges- genomic medicine, particularly if appropriate decision sup- tive of germline mutations, patients can be actively sought port tools present relevant data to clinicians only when needed. and counseled regarding the importance of genomic test- Patients genetically unable to activate prodrugs such as clopi- ing for their own treatment and for potential preventive care dogrel, tamoxifen, and codeine, for example, do not need their of their relatives. Direct outreach to patients at the Cleveland genetic data presented to their clinicians unless and until these Clinic, for example, increased the uptake of germline testing drugs are about to be prescribed. Although “reactive” genotyp- in patients with colorectal cancer by nearly sixfold—from 14% ing for pharmacogenomic variants can be ordered at many sites, with information provided directly to surgeons to 80% with a turnaround is often slow and uptake low. To address this, the genetic counselor seeing patients at follow-up examinations Vanderbilt University Medical Center and St. Jude Children’s (C.E., unpublished data). Highly penetrant variants such as Research Hospital have both implemented programs to con- these provide robust, evidence-based paradigms to help move duct preemptive genotyping in patients very likely to receive a clinical site toward implementing broader genomic medicine medications with relevant genetically based dosing algorithms. At Vanderbilt, for example, selected patients, including those Table 1 Examples of genomic medicine projects currently undergoing coronary angiography for acute coronary syn- in implementation at participating sites dromes receive genotyping on a 184-variant platform that Tumor-based genotype-driven treatment includes CYP2C19 alleles before being prescribed clopido- Risk/susceptibility testing in relatives of patients with -bearing grel,21 based on a review of available evidence of poorer out- cancer (Lynch syndrome, BRCA1/2, etc.) comes in persons with variant CYP2C19 alleles.22–24 At St. Jude Family history collection for assessment of individual risk Children’s Research Hospital, array-based testing for 225 CYP2C19 and antiplatelet therapy is performed, and results for those genes with the strongest Other genotype-driven treatment decisions clinical evidence are placed in the EMR.25 Clinical pharmacoge- Whole-genome/whole- sequencing for unknown disease diagnosis nomic testing is expanding to several sites affiliated with the Complex disease risk advice (myocardial infarction and type 2 diabetes) Pharmacogenomics Research Network (PGRN)26 with a goal of

Genetics in medicine | Volume 15 | Number 4 | April 2013 259 Review MANOLIO et al | Implementing genomic medicine in the clinic implementing proactive genotyping of a large number of vari- These four briefly described applications—tumor-based ants, such as those identified by PharmGKB27 or the US Food screening, family history–directed decision support, pharma- and Drug Administration (FDA).28 cogenomics, and diagnostic genome sequencing—demonstrate Preemptive testing, embedding this information into the that genomic medicine is no longer on the threshold; it has EMR with appropriate decision support, and then seamlessly arrived. Although much of this work is being pursued as part of presenting it at the point of care when needed, rather than research programs, some has been adopted institutionally and requiring tests to be ordered each time a relevant medication offered in a CLIA-certified laboratory environment as part of is prescribed, could help to minimize delay, cost, lack of fol- regular clinical care.13 Evaluating the impact of such programs low-through, and duplicate testing. This is a testable hypoth- and expanding their reach to diverse settings and populations esis and one that is likely to be explored in programs such as is a high priority within the National Human Genome Research the Electronic Medical Records and Genomics (eMERGE) Institute’s research agenda and mission. Information from such Network.29 Preemptive pharmacogenomic information could programs is helping to provide a needed evidence base to sup- be even more useful were informed patients able to access it port more widespread implementation and reimbursement. and work proactively with their clinicians in making treatment Consideration of the barriers encountered and strategies used decisions. Putting patients in control of their genomic informa- to surmount them by “early adopter” genomic medicine sites tion is another potential strategy for genomic medicine imple- may help speed their evaluation and facilitate their expansion mentation that needs evaluation. Such implementation strat- to the mainstream of clinical care. egies could be evaluated by measuring improvement, or lack thereof, in efficiency, cost, and health outcomes. CHALLENGES IN IMPLEMENTATION Whole-exome (protein-coding plus other highly con- Numerous challenges and barriers have been encountered in served regions) and whole-genome sequencing are beginning to launching genomic medicine projects and many common solu- be used for identifying genetic causes of rare, unknown condi- tions have emerged (Table 2). The greatest challenge may be the tions, particularly those presenting in childhood or in strongly lack of appreciation by clinicians, health-care institutions, and affected pedigrees for which novel mutations are more easily payers of the potential for genomics to improve patient care, detected.30 The recent cases of a young boy with severe, atypical, as compelling evidence of clinical validity and utility is limited refractory inflammatory bowel disease31 and of three families at present.37 Even for genomic applications with proven valid- with severe arterial calcifications32 demonstrate the potential ity and utility, such as family history, there is lack of adoption for this approach to detect novel disease-causing mutations and implementation, suggesting the lack of evidence is far from and point the way toward appropriate . The Medical being the only barrier. College of Wisconsin and Children’s Hospital of Wisconsin Cost concerns and institutional inertia typically demand have implemented a genome sequencing program to which cli- convincing arguments and hard data before clinical practice is nicians can nominate patients who remain undiagnosed after changed, yet the genome-wide nature of results to be expected appropriate clinical evaluation and testing. Nominations are by incorporating dense genotyping technologies into clini- reviewed by a multidisciplinary patient selection committee cal care will soon force clinicians and institutions to deal with chaired by the hospital’s chief medical officer. Patients and their genomic information for which evidence may be quite limited. families undergo 6–8 h of assessment and counseling before Evidentiary thresholds may thus need to be aligned with the enrollment. Sequencing is provided by a commercial labora- intended use of the information, as the impact of genotype- tory certified by the College of American Pathologists (CAP) driven care on morbidity and mortality cannot be tested for Laboratory Accreditation Program and the Clinical Laboratory every variant. Improvement Amendments (CLIA) Program. The detected Institutions have frequently relied on expert panels and local variants are analyzed using an in-house, validated software committees to evaluate available evidence and recommend par- package, focusing first on specific candidate genes nominated ticular new initiatives in genomic medicine, such as testing for by the referring clinicians to limit the chances of incidental, specific pharmacogenomic variants or estimating risk to carriers. potentially unrelated findings. Analysis is then broadened if ini- Such panels have tended to work in isolation from institution to tially nominated genes are uninformative. Potentially relevant institution, surveying the same evidence and often coming to variants are confirmed by an independent method, and those similar conclusions. Far better would be to harness the collective results are placed in the medical record. The Baylor College of knowledge of these groups in a more systematic way, distribut- Medicine has implemented a Whole Genome Laboratory33 with ing among them the genomic medicine issues for which evi- “in-house” sequencing and analysis of variants in a CLIA/CAP dence is compiled and evaluated, such as antidepressant therapy, environment with review and sign-out by a board-certified clin- coronary disease risk assessment, or risk in relatives of patients ical molecular . A number of other centers (Geisinger with colorectal cancer. These groups could then use a mutually Clinic, Partners Healthcare, Washington University) and com- accepted protocol for evaluation, with conclusions of each pro- mercial laboratories such as Ambry Genetics,34 GeneDx,35 and cess shared among all. This is similar to the approach taken by SeqWright36 are also now offering whole-genome sequencing the Centers for Disease Control and Prevention’s Evaluation of that can be used in clinical care. Genomic Applications in Practice and Prevention (EGAPP),38

260 Volume 15 | Number 4 | April 2013 | Genetics in medicine Implementing genomic medicine in the clinic | MANOLIO et al Review

Table 2 Challenges in implementation and potential solutions Challenge Potential solutions Limited Generate evidence of clinical utility of genomic medicine interventions evidence and Tailor needs for evidence against potential benefits and risks conflicting Convene expert panels to develop guidelines based on best available evidence interpretation of benefit/value Engage nonaffiliated community-based practices, where institutional resistance may be minimal, to assist in developing evidence Determine process outcomes of incorporating genomic information into EMRs Widely publicize successes Organize consortia to conduct multicenter trials of genotype-directed therapy when appropriate Lack of Establish institutional advisory committee(s) involving senior leadership to evaluate evidence, recommend, and monitor implementation institutional Engage early adopters and clinical champions in demonstration efforts and clinician Conduct pilot projects in early adopter clinical environments to develop results sufficient for follow-on funding acceptance Obtain transinstitutional commitment at highest levels involving all relevant departments and stakeholders Utilize internal pilot funding to catalyze initiation Build clinician acceptance of genetics professionals by judiciously integrating genetic counselors and/or in nongenetics clinical services throughout primary institution and affiliates Bring fragmented expertise for advancing genomic medicine under one transinstitutional center or institute Harness institutional quality improvement processes to assess and demonstrate value Limited access Use research screening assays on site and confirm clinically actionable findings with rapid, cost-effective CLIA-certified testing off-site if to genomic necessary medicine Establish or expand institutional CLIA-compliant genotyping to expand point-of-care testing, same-day service expertise and Choose platform to assay multiple important genotypes simultaneously, reliably, and cheaply testing Invest in new equipment and personnel to ensure research quality control is at the same level as the clinical laboratory; this requires an institutional investment Work with genetic counselors to establish protocol for process and parameters of data return Develop better justification and approaches for reimbursing efforts of clinicians and genetic counselors Assess and provide sufficient genetic counselor staff or ancillary staff Establish institutional genomic medicine interdisciplinary clinical case conference for discussion and guidance on difficult clinical cases Lack of standards Develop agreed-upon framework or standards for evaluation of genomic medicine applications for genomic Develop standardized order sets and process modification applications Develop standards for analytic validity of whole-genome and whole- sufficient for clinical application, and for interpretation of the variants found by these methods EMR integration Enable access to actionable genetic information in the EMR through development of user-friendly decision support algorithms for of genomic health-care providers results and Establish “Usability Lab” to assess CDS tools and minimize alert fatigue clinical decision Allocate genomic medicine institutional funding to develop education and outreach to disseminate “best practices” incorporating family support history and genomic information Redesign the EMR to include section dedicated to containing all relevant genotype results for a patient Develop and link actionable drug– pair decision support to electronic pharmaceutical ordering software at point of care to facilitate adoption of pharmacogenomic testing Establish interdisciplinary workgroup with genomic medicine and EMR team to create secure tools for EMR-based genomic decision support Follow-up of Shift from relying on primary physician contact to direct contact and follow-up by genomic medicine team, with permission of physician genotyped Analyze and address reasons for refusal to complete confirmatory testing, such as lack of for testing patients Outreach to Clarify implications for family members and health-care providers’ responsibilities towards family members at-risk family Explore ways to improve information to at-risk family members members Consent Ensure consents for implementation projects conducted as research studies include returning results to patients and entering results into medical record Conformance with standard of care and specific consent may not be needed Consider initial implementation projects that might not require consent, such as results in established clinical pathways (tumor mutations, CYP2C19*2) Develop standards for informed consent for extensive sequencing, including whole-genome sequencing, and obtain it prospectively Ensure that availability of personnel to manage consent/counseling is not rate limiting in initial implementation CDS, clinical decision support; CLIA, Clinical Laboratory Improvement Amendments; EMR, electronic medical record. Table 2 Continued on next page

Genetics in medicine | Volume 15 | Number 4 | April 2013 261 Review MANOLIO et al | Implementing genomic medicine in the clinic

Table 2 Continued Challenge Potential solutions Understanding Conduct focus groups of patients, clinicians, and ancillary personnel to identify specific educational needs by patients, Conduct genetic and genomic medicine education campaign for patients, clinicians, and ancillary personnel based on focus group input clinicians, public Survey retention of educational information by patients and clinicians and modify programs as needed Conduct genotyping and/or sequencing and interpretation exercises with medical and other health profession students Provide specific health-care provider education on when to order test, and how to interpret and how to act on implications for family members and provider’s responsibilities Introduce pharmacogenomic lectures into health professionals’ training and continuing education Provide clinical supervision to clinician trainees in use of pharmacogenomic testing, other genomic point-of-care testing Lack of access Combine current centers’ small patient collections of “normal” sequences and make available to all to comparison Prioritize funding for costly and time-consuming storage of viable tissues “control” Bank high-quality tumor samples for confirmatory clinical sequencing with patient identifiers sequence data and banking High-level institutional commitment to combine and organize multiple biorepositories for efficiency and ease of access resources Lack of research Until evidence base is established for making genomic testing new standard of care, consider research funding for testing in interim funding and between discovery and adoption reimbursement Gradually change culture to convince health-care community and patients of value of genomic medicine and need for reimbursement Demonstrate cost of testing is not prohibitive and savings and impact can be substantial Provide institutional back-up for reimbursement to avoid charges to patients Anticipate rises in interpretive and delivery costs as technology cost drops and enthusiasm increases CDS, clinical decision support; CLIA, Clinical Laboratory Improvement Amendments; EMR, electronic medical record. the Clinical Pharmacogenomics Implementation Consortium,39 over 40 genetic loci, for example, whole-exome sequencing can and by some early adopter institutions in developing commit- already be performed for less than half the price of the current tees of their own. Given the rapid spread of diagnostic sequenc- clinically available “multigene panel.”42 Third-party reimburse- ing and direct-to-consumer testing, it may be appropriate to ment of these costs is thus crucial for ensuring uptake by insti- begin to identify an intermediate category of “actionable” vari- tutions and patients. ants, for which evidence is insufficient to establish unequivocal Integrating effective clinical decision support tools into the clinical utility but is sufficient to determine how already avail- EMR, to query a patient’s genomic data as care is being delivered, able information could be used. Consideration should be given will be needed to present relevant variants and recommenda- to tailoring of required levels of evidence to the risk/benefit ratio tions only when they are indicated. Such real-time, genomically of the specific clinical setting at issue; development of consen- driven clinical decision support can help avoid “alert fatigue” sus on evidence needed in different scenarios would be a valu- and relieve the clinician of trying to sift through the massive able contribution. Recommendations for drastic or irreversible amount of genomic information without impeding clinical interventions such as genotype-driven mastectomy or salpingo– workflow. Sharing the effort of following up actionable genetic oophorectomy, for instance, should require far stricter evidence variants and at-risk family members with a dedicated genomic than should low-risk interventions such as modifying choice medicine team, with the primary physician’s agreement and between two proven, roughly equivalent medications such as in ways that do not “compete” for the patient, can also mini- codeine and hydromorphone.40 mize burden on patients and clinicians and increase efficiency. For more modest interventions, the risks are more likely Looking to the future, the potential to update EMRs automati- to involve factors such as burden to clinicians and patients in cally as new discoveries are made and their clinical utility vali- obtaining, interpreting, and managing the result, including dated provides exciting possibilities for rapid implementation learning of incidental findings not sought by clinician or patient; of research advances. Growing knowledge and understanding, cost of testing; and psychological impact of the information on however, mean that variants that are today of uncertain impor- patients or their families. The first two of these can be addressed tance may tomorrow be found to be of great clinical relevance in part by improving the accessibility and lowering the costs of and vice versa. The responsibility of the clinician to recontact obtaining genomic information suitable for clinical use, which the patient with this new information is uncertain and could is occurring rapidly with expanding certification of genomic pose a major burden as well as medico-legal risks.43 technologies and laboratories under CLIA and CAP. Testing Psychological impact on patients and their families has tra- fees can be substantial, especially when individual genetic vari- ditionally been addressed by the informed consent process and ants are assayed sequentially, and indeed are soon likely to rival , but expanding genomic medicine efforts the cost of obtaining an entire genome sequence.41 For a geneti- will likely soon outstrip the available supply of genetic counsel- cally heterogeneous condition, such as Charcot-Marie-Tooth ors.44 Innovative models utilizing ancillary personnel or social disease, that is potentially due to variation at one or more of media tools to provide counseling and education, along with

262 Volume 15 | Number 4 | April 2013 | Genetics in medicine Implementing genomic medicine in the clinic | MANOLIO et al Review broad-ranging educational campaigns to increase familiarity openly accessible to clinical groups attempting to interpret with genomic medicine among practitioners, patients, and pay- sequence data. Such a knowledge base would have enormous ers, will have the greatest impact if they can be linked to early value for research as well, particularly to the degree that it successes and readily observable local efforts. entails a careful curation and consensus process to reduce error Reports of successes such as reduced hospitalization rates and promote consistency. Here again, many genomic medi- following genotype-driven adjustments in warfarin dosing,45 cine groups appear to be doing this in isolation, capturing and e.g., or symptomatic improvement or reduced adverse effects reviewing such information internally but finding themselves after genotype-driven selection of antidepressants,46 can help constrained in sharing it outside their institutions. The policy stimulate discussion of the potential for genomic medicine to aspects of establishing such a knowledge base, ensuring the improve outcomes and shift an institution’s culture to embrace accuracy of associations to be reported within it, and providing genotype-guided care,30 although even these examples need a wide access to it are far from trivial and involve patient privacy more complete evidence base and remain controversial.47–50 and confidentiality of health information. Linking to appropri- ate supporting evidence and providing frequent updates are also INFRASTRUCTURE AND RESEARCH NEEDS daunting tasks, given the rapid pace of discovery, but are tasks Genomic medicine programs share needs for basic informa- that could readily be distributed among collaborating groups. tional and policy infrastructure as well as evidence and out- Perhaps an even greater informational need addressable comes produced through research (Table 3). Key infrastructure by data sharing among genomic medicine groups and other needs include a comprehensive knowledge base that captures genome sequencing projects is the ability to determine whether sequence variants and their phenotypic associations and is a newly discovered variant has been reported before. This is

Table 3 Infrastructure and research needs Infrastructure Continuously updated knowledge base and look-up tables of sequence variants linked to clinical phenotypes, genotype–phenotype relationships, actionable variants, and supportive evidence Reliable, standardized genotyping/sequencing platforms, reporting formats, and quality control High-throughput, rapid, low-cost genotyping/sequencing in CAP/CLIA-certified environments, potentially concentrated in large service centers for sequencing and interpretation Evidentiary standards for various types of genomic testing in collaboration with other agencies (FDA, CMS) as needed, including standards and criteria for reimbursement Point-of-care educational information and clinical decision support Genomic medicine training programs for , pharmacists, nurses, and other clinicians Slide sets, videos, seminars, and social media approaches to patient, public, and clinician education Increased person-power: clinicians with genetics training and background, genomic medicine consultants, genetic counselors, and implementation scientists Expanded EMR capabilities to incorporate family history information, interface with genomic data Research Evidence development, such as changes following genomic medicine implementation in: Disease severity, progression, or incidence At-risk persons (such as family members) identified Patients’ or family members’ anxiety, quality of life, and access to care or insurance Cost, compliance, and patient or clinician satisfaction Comparative effectiveness research Identification of actionable variants Bioinformatic/analysis tools to facilitate making associations, identifying “actionability” or impact scores for specific variants Procedures and resources for interpreting potential clinical relevance of “variants of unknown significance,” such as clinical proteomic tests, RNA-splicing tests, and cell-based assays, to determine experimentally whether a novel variant might cause disease Standardized consent permitting return of results, deposition in EMR, sharing of variant tables Collaborative projects/network to pool resources, share approaches, identify best practices, and avoid duplication Best practices for: Implementation and dissemination Defining and collecting outcomes of implementation, including process outcomes Delivering WGS information to individual providers and patients Clearinghouse of successful implementation projects CAP, College of American Pathologists; CLIA, Clinical Laboratory Improvement Amendments; CMS, Centers for Medicare and Medicaid Services; EMR, electronic medical record; FDA, Food and Drug Administration; WGS, whole-genome sequencing.

Genetics in medicine | Volume 15 | Number 4 | April 2013 263 Review MANOLIO et al | Implementing genomic medicine in the clinic currently done by surveying available large-scale databases, such earlier in their training.60 Fellowships in pharmacogenomics as the 1000 Project,51 the Exome Variant Server,52 and and genomic medicine will be needed to develop specialists local institutional catalogs, on the presumption that a patient (perhaps “clinical genomicists,” akin to radiologists or patholo- with a very rare or undiagnosed condition likely has a private gists) capable of interpreting patients’ genomic information or even unique variant.53 Interpretation can be difficult because and advising clinicians on appropriate actions to be taken for a these reference databases generally lack phenotypic information. given set of variants in a given clinical setting. Given the broad Far better than individual small cohorts of sequenced patients range of possible genomic medicine applications, the training would be to aggregate cohorts and make their data widely avail- of physicians and scientists enrolled in certified programs of able to qualified clinicians and researchers. Sequence variants in the American Board of and Genomics and these databases should also be linked to phenotypic informa- the American Board of in clinical genetics, molecu- tion, when possible, rather than forcing the assumption that all lar genetics, biochemical genetics, and is particu- available sequenced individuals are free of the disease under larly relevant. Consideration should be given to the potential consideration. The recently initiated Centers for Mendelian for expanding the curricula of these programs to include more Genomics54,55 will identify variants responsible for a great num- genomic medicine content or providing fellowships within ber of rare disorders and make this information widely available, these specialties that focus on genomic medicine. which should help in this regard. The nascent ClinVar database Other educational modes such as seminars, slide sets, webi- of the National Center for Biotechnology Information56 and the nars, and videos need to be developed and shared across insti- International Standards for Cytogenomic Arrays database avail- tutions doing similar types of work. Expanded EMR capabili- able in dbGaP and dbVar57 are two open-access databases that ties to permit querying of sequence or other dense genomic can provide valuable phenotypic information associated with data, and incorporation of family history, are readily address- rare sequence variants and structural variation. able. EMRs also have the ability to deliver context-specific edu- Placing all this information in a single queryable source, cational information at the point of care during an eminently regardless of where the actual data reside, would be a substan- teachable moment, with a clinical conundrum at hand, as well tial advance in interpreting the three million or more differences as more sophisticated clinical decision support to aid clinicians from the human reference sequence identifiable per individual.58 in the appropriate use of genomic information.61 Challenges in Genomic sequencing, in contrast to genotyping microarray implementing point-of-care tools need to be recognized, how- technology, reveals rare or even unique alleles lacking scien- ever, including the potential for overwhelming clinicians with tific literature support for physiological effects. Computational alerts leading to alert fatigue.62 Here again, consideration should approaches to predicting function may need to be incorporated be given to the evidence needed for implementing such tools in into clinical decision support methods in trying to determine genomic medicine and for ensuring the consistent adoption of the implications of such variants for clinical care. Such analyses point-of-care interventions for high priority health-care targets would be greatly facilitated by standard formats for reporting with the best risk/benefit ratio and strongest evidence, whether the data produced by individual sequencing groups rather than genomic or nongenomic. expecting individual clinicians and clinical sites to handle and Research is acutely needed to generate, collect, and make interpret multiple different formats. The Centers for Disease widely available the evidence needed for determining which Control and Prevention are currently working to develop perfor- variants are actionable, in whom, and in what clinical situa- mance specifications for next-generation sequencing.59 Access to tions. Intermediate steps are needed that lie between genomic accurate, low-cost genome sequencing in CAP/CLIA-certified discovery research and routine clinical implementation to gen- environments will remain a major limiting factor in utilizing erate the evidence base to justify implementation. Consensus sequencing in clinical care across diverse settings. Provision should be sought on the levels of evidence needed to justify the of genome sequencing as a commodity service in national or introduction of genomic medicine interventions, taking into regional hubs, including interpretation of potential clinical rel- account the degree of complexity or cost (such as family his- evance of identified variants, could be a major advance. tory vs. whole-genome sequencing) and risk/benefit (Lynch Appropriate professional organizations that can develop syndrome screening vs. universal colonscopy) involved, and the genomic medicine practice guidelines, such as the American comparisons and outcome measures needed to answer these College of Medical Genetics and Genomics, the College of questions. Generating clear evidence of benefit for genomic American Pathologists, and the Association of Molecular interventions will remain challenging, however, in part because Pathologists, as well as regulatory agencies such as the FDA the rarity of many genomic markers of risk means large samples and Centers for Medicare and Medicaid Services, should be must be studied to accrue benefit/risk data in the small sub- involved in developing evidentiary standards tailored to spe- sets of patients affected. Even with large samples, there can be cific interventions under consideration and criteria for reim- controversy about the interpretation of the evidence, as shown bursement. Targeted education for active practitioners should by the data on the relationship between CYP2C19 variants and also be tailored to specific settings and delivered as succinctly failure of antiplatelet efficacy of clopidogrel.63 In one meta- as possible at the point of care for maximum uptake and value, analysis that included only patients with acute coronary syn- whereas broader educational efforts can capture clinicians dromes, there was strong evidence for an effect of variant alleles

264 Volume 15 | Number 4 | April 2013 | Genetics in medicine Implementing genomic medicine in the clinic | MANOLIO et al Review on outcomes,23 whereas another analysis that included trials to convincing the various stakeholders to accept an imple- with different indications (and in which the efficacy of clopido- mentation project, particularly the first one (Figure 1). Such grel itself appears to be less) reported a much smaller effect of grounding can be obtained from a comprehensive review of on outcome.24 Thus, the evidence supporting the available scientific literature, recommendations of expert an effect of pharmacogenomic variants on variable drug out- groups, and examination of ongoing successful projects at comes may also depend on indication, substantially increasing other institutions. Available policy or institutional guidelines the complexity of generating and evaluating evidence. on implementing changes to clinical care must also be consid- Research efforts should also aim to establish and validate cri- ered. Critical in choosing an initial implementation project is teria for actionability. They should facilitate association analysis the pragmatic issue of identifying a group of ready adopters or and assessment of actionability and influence on outcomes from clinical champions, such as a group or local prac- available data. Given that it is impossible, however, to perform tice, willing to try a new approach. Indeed, selection of a pilot prospective studies of the impact of all potentially important project among several possibilities may hinge upon the avail- variants on clinical outcomes, comprehensive strategies will be ability of an enthusiastic group of clinicians. needed to generate the necessary data. Real-world studies quan- Identifying and engaging the myriad stakeholders within tifying the impact of implementing genomic medicine programs an institution, including those needed to conduct genetic test- on patient outcomes, cost and acceptability of care, and clini- ing, interpret it, integrate it with the EMR, provide results and cian and patient satisfaction are needed to promote adoption recommendations to the clinician, and pay for it, require that and dissemination of genomic medicine. Such studies fall under senior leaders who are familiar with the institution and its cul- the rubric of “implementation research,” defined as the study of ture be on the genomic medicine team. It also often requires methods to promote the systematic uptake of proven interven- patience, persistence, and sufficient standing within the organi- tions into routine practice.64 Comparative effectiveness research zation to convince skeptics and overcome institutional inertia. assessing the added value of genomic information, such as Some programs have found it simpler to begin at smaller affili- genomic variants added to family history added to clinical risk ated hospitals or practices rather than at dominant academic factors, and consensus on the outcomes and evidence needed to medical centers. answer such questions, would help to define priorities for col- Engagement and support from institutional leadership is lecting and using such information in a cost-effective way. obviously crucial, and most successful programs have involved Informed consents for genomic research should be broad- ened to include return of results, where desired and appropriate, Scientific evidence, experience, unmet need Local Available so that lack of such consent (or indeed, consent forms declaring champions policy or or ready suitable “no results will ever be returned”) does not become an obstacle, Literature Successful adopters guidelines review external projects if not an excuse, for failing to report actionable findings. The consent process should also explicitly address deposition of data in the EMR or research databases and sharing of variant Select pilot project information among researchers and clinicians. Identify and engage Engage institutional Assembling early adopters of genomic medicine into a confed- stakeholders leadership eration or community should facilitate addressing the many infra- structure and research needs outlined here. Collaborative projects Secure funding such as those of the PGRN65 can facilitate dissemination and test- ing of genomic medicine approaches in diverse settings to maxi- Publicize, educate patients, mize their generalizability and usefulness. Such projects can help clinicians develop best practices for complex tasks such as implementation and dissemination, defining and collecting outcomes of imple- Develop, pilot clinical mentation, and delivering genomic information to clinicians and workflow patients. A clearinghouse of successful implementation projects, with detailed protocols addressing steps needed for patients, clini- Launch, monitor initial cases for protocol cians, laboratories, departments, and institutions, would go a long adherence, problems way toward disseminating this work beyond early adopter sites.

Indeed, a critical question to ask is whether and how approaches Collect outcome data, developed at highly specialized and resourced tertiary care centers modifiy protocol as needed can be adopted in less resource-intensive settings. Disseminate, expand to IMPLEMENTATION ROADMAP partner sites

Incorporating genomic results into clinical care is as much a Figure 1 Implementation roadmap: one approach to initial imple- cultural and political exercise within a given institution as a mentation at a single institution with subsequent evaluation and scientific one, although a firm scientific grounding is essential ultimately dissemination.

Genetics in medicine | Volume 15 | Number 4 | April 2013 265 Review MANOLIO et al | Implementing genomic medicine in the clinic them directly in decision making, as chairs of committees disease. It is now time to ensure those results are translated selecting interventions to implement, for example, or evalu- and used to the maximum benefit of patients by periodically ating their impact. Funding may be provided through local convening relevant stakeholders and building on the vanguard institutional support, in anticipation perhaps of eventual reim- of nationwide genomic medicine implementation projects. bursement by major payers, or through foundation or research Bringing these groups together in this way is anticipated to funding, including National Institutes of Health–supported increase their diffusion, efficiency, and, ultimately, their value projects such as the Clinical Translational Awards and to patients and health care. the PGRN. Institutional stakeholders are key in developing practical and efficient clinical workflows for obtaining testing ACKNOWLEDGMENTS and providing results. Development and testing of streamlined This article summarizes the deliberations of a colloquium con- educational tools, which may be as simple as a pop-up window vened by the National Human Genome Research Institute on with appropriate links for the clinician and letters or brochures 29 June 2011 to examine opportunities and barriers for imple- for patients, will be essential both for initiating a program and menting genomic medicine in clinical care. The authors gratefully encouraging its acceptance. Once launched, careful moni- acknowledge the contributions of Gary Gibbons to the workshop. toring of initial cases for adherence to established workflows and unanticipated barriers permits tailoring of procedures as DISCLOSURE needed. Outcome data related to ease of use, adherence, patient C.E. is an unpaid member of the External Scientific Advisory Board and clinician satisfaction and behaviors, cost, and, where possi- of MyMercury.com and of the Genomic Medicine Advisory Board ble, measures of morbidity and mortality can be used to modify of Complete Genomics, Inc. The other authors declare no conflict the program and inform design and implementation of future of interest. efforts. Evaluation of the effectiveness of an implementation project should focus on outcomes of the implementation itself REFERENCES rather than the outcomes of treatment.66 Several conceptual 1. Dulbecco R. A turning point in cancer research: sequencing the human genome. 67,68 Science 1986;231:1055–1056. models for such evaluations are available for consideration. 2. Collins FS. Shattuck lecture–medical and societal consequences of the Human . N Engl J Med 1999;341:28–37. CONCLUSION 3. Guttmacher AE, Collins FS. Genomic medicine–a primer. N Engl J Med 2002;347:1512–1520. Multiple institutions are actively engaged in using patients’ 4. Evans JP, Meslin EM, Marteau TM, Caulfield T. Genomics. Deflating the genomic genomic information in their clinical care, either as pilot dem- bubble. Science 2011;331:861–862. onstration and research projects or as part of clinical care as 5. Varmus H. Ten years on–the human genome and medicine. N Engl J Med 2010;362:2028–2029. a reimbursable service. The Genomic Medicine Colloquium 6. Committee on Quality of Health Care in America, Institute of Medicine. revealed that much of this work is being done in isolation and Crossing the Quality Chasm: A New Health System for the 21st Century. would benefit from more structured collaboration and sharing National Academy Press: Washington, DC, 2001. of approaches across groups and institutions. Implementation 7. England JD, Gronseth GS, Franklin G, et al.; American Academy of . Practice Parameter: evaluation of distal symmetric polyneuropathy: role of of genomics in medicine would also benefit from establishing laboratory and (an evidence-based review). Report of the common infrastructure, such as a catalog of variants observed American Academy of Neurology, American Association of Neuromuscular and across large numbers of sequenced individuals to date, an Electrodiagnostic Medicine, and American Academy of Physical Medicine and Rehabilitation. Neurology 2009;72:185–192. updatable database of actionable variants, evidentiary stan- 8. Lupski JR, de Oca-Luna RM, Slaugenhaupt S, et al. DNA duplication associated dards tailored to benefits and risks, common clinical decision with Charcot-Marie-Tooth disease type 1A. Cell 1991;66:219–232. support tools including point-of-care educational materials, 9. Manolio TA, Brooks LD, Collins FS. A HapMap harvest of insights into the genetics of common disease. J Clin Invest 2008;118:1590–1605. and collaborative projects to pool resources and identify best 10. Ladabaum U, Wang G, Terdiman J, et al. Strategies to identify the Lynch practices. Systematic and timely incorporation of advances in syndrome among patients with colorectal cancer: a cost-effectiveness analysis. genomic research related to somatic and germline DNA, as well Ann Intern Med 2011;155:69–79. as consideration of RNA, protein, and epigenomic applications, 11. Coates R, Williams M, Melillo S, Gudgeon J. Genetic testing for lynch syndrome in individuals newly diagnosed with colorectal cancer to reduce morbidity and will also be needed. The National Human Genome Research mortality from colorectal cancer in their relatives. PLoS Curr 2011;3:RRN1246. Institute, in collaboration with several other National Institutes 12. US Preventive Services Task Force. Genetic risk assessment and BRCA mutation of Health institutes and in accord with its recent strategic plan,69 testing for breast and ovarian cancer susceptibility: recommendation statement. Ann Intern Med 2005;143:355–361. is currently exploring approaches for facilitating the research 13. Gudgeon JM, Williams JL, Burt RW, Samowitz WS, Snow GL, Williams MS. needed to implement genomic medicine on an ever-widening Lynch syndrome screening implementation: business analysis by a healthcare scale. Such efforts should also take full advantage of the extant system. Am J Manag Care 2011;17:e288–e300. 14. Qureshi N, Armstrong S, Dhiman P, et al.; for the ADDFAM (Added Value of momentum, information, and experience of “early adopters” Family History in CVD Risk Assessment) Study Group. Effect of adding systematic across the United States, as well as the tools of implementation family history enquiry to cardiovascular disease risk assessment in primary care: science to address the cultural barriers described above. A Matched-Pair, Cluster Randomized Trial. Ann Intern Med 2012;156:253–262. 15. Rubinstein WS, Acheson LS, O’Neill SM, et al.; Family Healthware Impact Trial The substantial National Institutes of Health investment in (FHITr) Group. Clinical utility of family history for cancer screening and referral basic genomic research has yielded and will continue to yield in primary care: a report from the Family Healthware Impact Trial. Genet Med new insights into the role of genetic variants in health and 2011;13:956–965.

266 Volume 15 | Number 4 | April 2013 | Genetics in medicine Implementing genomic medicine in the clinic | MANOLIO et al Review

16. Khoury MJ, Feero WG, Valdez R. Family history and as tools 44. National Cancer Institute, Research-tested Intervention Programs. http://rtips. for improving health in an era of evidence-based medicine. Am J Prev Med cancer.gov/rtips/programDetails.do?programId=277363. 2010;39:184–188. 45. Wang L, McLeod HL, Weinshilboum RM. Genomics and drug response. N Engl J 17. Surgeon General’s My Family Health Portrait. https://familyhistory.hhs.gov/fhh- Med 2011;364:1144–1153. web/home.action. 46. Porcelli S, Drago A, Fabbri C, Gibiino S, Calati R, Serretti A. Pharmacogenetics of 18. Orlando LA, Hauser ER, Christianson C, et al. Protocol for implementation of antidepressant response. J Psychiatry Neurosci 2011;36:87–113. family health history collection and decision support into primary care using a 47. Verhoef TI, Schalekamp T, Redekop WK, de Boer A, Maitland-van der Zee AH. computerized family health history system. BMC Health Serv Res 2011;11:264. Clinical and economic consequences of pharmacogenetic-guided dosing of 19. Doerr M, Leach B, Men X, et al. Patient-entered, web-based, EMR-compatible, warfarin. Expert Rev Pharmacoecon Outcomes Res 2010;10:375–378. comprehensive cancer family history collection tool: clinical workflow, 48. Meckley LM, Gudgeon JM, Anderson JL, Williams MS, Veenstra DL. A policy demographics, and patient satisfaction from pilot study. American Society of model to evaluate the benefits, risks and costs of warfarin pharmacogenomic Human Genetics 2010 Annual Meeting: Abstract 1992. Washington, D.C., 2–6 testing. Pharmacoeconomics 2010;28:61–74. November 2010. 49. Evaluation of Genomic Applications in Practice and Prevention (EGAPP) Working 20. Hulse NC, Ranade-Kharkar P, Post H, Wood GM, Williams MS, Haug PJ. Group. Recommendations from the EGAPP Working Group: testing for Development and early usage patterns of a consumer-facing family health polymorphisms in adults with nonpsychotic depression treated history tool. AMIA Annu Symp Proc 2011;2011:578–587. with selective serotonin reuptake inhibitors. Genet Med 2007;9:819–825. 21. Pulley JM, JC, Peterson JF, et al. Operational implementation of 50. Anderson JL, Horne BD, Stevens SM, et al. A randomized and clinical effectiveness prospective genotyping for personalized medicine: the design of the Vanderbilt trial comparing two pharmacogenetic algorithms and standard care for PREDICT project. Clin Pharmacol Ther 2012;92:87–95. individualizing warfarin dosing (CoumaGen-II). Circulation 2012;125:1997–2005. 22. Collet JP, Hulot JS, Pena A, et al. Cytochrome P450 2C19 polymorphism in 51. . http://www.1000genomes.org/. young patients treated with clopidogrel after myocardial infarction: a cohort 52. Exome Variant Server. http://evs.gs.washington.edu/EVS/. study. Lancet 2009;373:309–317. 53. Lupski JR, Belmont JW, Boerwinkle E, Gibbs RA. Clan genomics and the complex 23. Mega JL, Simon T, Collet JP, et al. Reduced-function CYP2C19 genotype and architecture of human disease. Cell 2011;147:32–43. risk of adverse clinical outcomes among patients treated with clopidogrel 54. Mendelian Disorders Sequencing Centers. http://www.genome.gov/27546192. predominantly for PCI: a meta-analysis. JAMA 2010;304:1821–1830. 55. Bamshad MJ, Shendure JA, Valle D, et al.; Centers for Mendelian Genomics. 24. Holmes MV, Perel P, Shah T, Hingorani AD, Casas JP. CYP2C19 genotype, The Centers for Mendelian Genomics: a new large-scale initiative to identify clopidogrel metabolism, platelet function, and cardiovascular events: a the genes underlying rare Mendelian conditions. Am J Med Genet A systematic review and meta-analysis. JAMA 2011;306:2704–2714. 2012;158A:1523–1525. 25. St. Jude Children’s Research Hospital PG4KDS program. http://www.stjude.org/ 56. National Center for Biotechnology Information. http://www.ncbi.nlm.nih.gov/ pg4kds. clinvar/. 26. Pharmacogenomics Research Network (PGRN). http://www.nigms.nih.gov/ 57. International Standards for Cytogenomic Arrays (ISCA) database. http://www. Research/FeaturedPrograms/PGRN/. iscaconsortium.org. 27. Pharmacogenomics Knowledge Base (PharmGKB). http://www.pharmgkb.org/ 58. Bainbridge MN, Wiszniewski W, Murdock DR, et al. Whole-genome sequencing search/guidelineList.action. for optimized patient management. Sci Transl Med 2011;3:87re3. 28. US Food and Drug Administration. Table of Pharmacogenomic Biomarkers 59. Centers for Disease Control and Prevention, Next Generation Sequencing: in Drug Labels. http://www.fda.gov/drugs/scienceresearch/researchareas/ Standardization of Clinical Testing (Nex-StoCT) Working Group. http://www. pharmacogenetics/ucm083378.htm. cdc.gov/osels/lspppo/Genetic_Testing_Quality_Practices/Nex-StoCT.html. 29. McCarty CA, Chisholm RL, Chute CG, et al.; eMERGE Team. The eMERGE 60. Dhar SU, Alford RL, Nelson EA, Potocki L. Enhancing exposure to genetics Network: a consortium of biorepositories linked to electronic medical records and genomics through an innovative curriculum. Genet Med data for conducting genomic studies. BMC Med Genomics 2011;4:13. 2012;14:163–167. 30. Gonzaga-Jauregui C, Lupski JR, Gibbs RA. Human genome sequencing in 61. Hoffman MA, Williams MS. Electronic medical records and personalized health and disease. Annu Rev Med 2012;63:35–61. medicine. Hum Genet 2011;130:33–39. 31. Worthey EA, Mayer AN, Syverson GD, et al. Making a definitive diagnosis: 62. Shah NR, Seger AC, Seger DL, et al. Improving acceptance of computerized successful clinical application of whole exome sequencing in a child with prescribing alerts in ambulatory care. J Am Med Inform Assoc 2006;13:5–11. intractable inflammatory bowel disease. Genet Med 2011;13:255–262. 63. Johnson JA, Roden DM, Lesko LJ, Ashley E, Klein TE, Shuldiner AR. Clopidogrel: 32. St Hilaire C, Ziegler SG, Markello TC, et al. NT5E mutations and arterial a case for indication-specific pharmacogenetics. Clin Pharmacol Ther calcifications. N Engl J Med 2011;364:432–442. 2012;91:774–776. 33. Baylor College of Medicine, Whole Genome Laboratory. http://www.bcm.edu/ 64. Eccles MP, Foy R, Sales A, Wensing M, Mittman B. Implementation Science six geneticlabs/index.cfm?PMID=21319. years on–our evolving scope and common reasons for rejection without review. 34. Ambry Genetics. http://www.ambrygen.com/clinical-diagnostic-exome. Implement Sci 2012;7:71. 35. GeneDx. http://www.genedx.com/site/xomedx. 65. Pharmacogenomics Research Network Research Groups. http://pgrn.org/ 36. SeqWright. http://www.seqwright.com/Exome. display/pgrnwebsite/Research+Groups+and+Network+Resources. 37. Scheuner MT, Sieverding P, Shekelle PG. Delivery of genomic medicine for 66. Proctor E, Silmere H, Raghavan R, et al. Outcomes for implementation research: common chronic adult diseases: a systematic review. JAMA 2008;299:1320– conceptual distinctions, measurement challenges, and research agenda. Adm 1334. Policy Ment Health 2011;38:65–76. 38. Teutsch SM, Bradley LA, Palomaki GE, et al.; EGAPP Working Group. The 67. Proctor EK, Landsverk J, Aarons G, Chambers D, Glisson C, Mittman B. Evaluation of Genomic Applications in Practice and Prevention (EGAPP) Implementation research in mental health services: an emerging science with Initiative: methods of the EGAPP Working Group. Genet Med 2009;11:3–14. conceptual, methodological, and training challenges. Adm Policy Ment Health 39. Relling MV, Klein TE. CPIC: Clinical Pharmacogenetics Implementation 2009;36:24–34. Consortium of the Pharmacogenomics Research Network. Clin Pharmacol Ther 68. Gaglio B, Glasgow RE. Evaluation approaches for dissemination and 2011;89:464–467. implementation research. In: Brownson RC, Colditz GA, Proctor EK (eds.), 40. Crews KR, Gaedigk A, Dunnenberger HM, et al.; Clinical Pharmacogenetics Dissemination and Implementation Research in Health: Translating Science to Implementation Consortium. Clinical Pharmacogenetics Implementation Practice. Oxford University Press, New York, NY 2012, pp. 334–337. Consortium (CPIC) guidelines for codeine therapy in the context of cytochrome 69. Green ED, Guyer MS; National Human Genome Research Institute. P450 2D6 (CYP2D6) genotype. Clin Pharmacol Ther 2012;91:321–326. Charting a course for genomic medicine from base pairs to bedside. Nature 41. Salzberg SL, Pertea M. Do-it-yourself genetic testing. Genome Biol 2011;470:204–213. 2010;11:404. 42. Lupski JR, Reid JG, Gonzaga-Jauregui C, et al. Whole-genome sequencing This work is licensed under a Creative in a patient with Charcot-Marie-Tooth neuropathy. N Engl J Med 2010;362: Commons Attribution-NonCommercial-No 1181–1191. 43. Pyeritz RE. The coming explosion in genetic testing–is there a duty to recontact? Derivative Works 3.0 License. To view a copy of this license, N Engl J Med 2011;365:1367–1369. visit http://creativecommons.org/licenses/by-nc-nd/3.0/

Genetics in medicine | Volume 15 | Number 4 | April 2013 267